Research article
Climate change and the economics of biomass energy feedstocks in semi-arid agricultural landscapes: A spatially explicit real options analysis

https://doi.org/10.1016/j.jenvman.2017.01.049Get rights and content

Highlights

  • We examine land use change from agriculture to biomass using real option analysis.

  • We consider impacts of climate change on spatially varying production risks.

  • Interacting, multiple uncertainties create option value, delaying land use change.

  • We show that these values can be substantial, creating a risk hurdle or premium.

  • Premiums differ and change with spatially-varying risks and climate change.

Abstract

The economics of establishing perennial species as renewable energy feedstocks has been widely investigated as a climate change adapted diversification option for landholders, primarily using net present value (NPV) analysis. NPV does not account for key uncertainties likely to influence relevant landholder decision making. While real options analysis (ROA) is an alternative method that accounts for the uncertainty over future conditions and the large upfront irreversible investment involved in establishing perennials, there have been limited applications of ROA to evaluating land use change decision economics and even fewer applications considering climate change risks. Further, while the influence of spatially varying climate risk on biomass conversion economic has been widely evaluated using NPV methods, effects of spatial variability and climate on land use change have been scarcely assessed with ROA. In this study we applied a simulation-based ROA model to evaluate a landholder's decision to convert land from agriculture to biomass. This spatially explicit model considers price and yield risks under baseline climate and two climate change scenarios over a geographically diverse farming region. We found that underlying variability in primary productivity across the study area had a substantial effect on conversion thresholds required to trigger land use change when compared to results from NPV analysis. Areas traditionally thought of as being quite similar in average productive capacity can display large differences in response to the inclusion of production and price risks. The effects of climate change, broadly reduced returns required for land use change to biomass in low and medium rainfall zones and increased them in higher rainfall areas. Additionally, the risks posed by climate change can further exacerbate the tendency for NPV methods to underestimate true conversion thresholds. Our results show that even under severe drying and warming where crop yield variability is more affected than perennial biomass plantings, comparatively little of the study area is economically viable for conversion to biomass under $200/DM t, and it is not until prices exceed $200/DM t that significant areas become profitable for biomass plantings. We conclude that for biomass to become a valuable diversification option the synchronisation of products and services derived from biomass and the development of markets is vital.

Introduction

De-carbonising global electricity generation is seen as key to stabilise atmospheric greenhouse gas levels (Edenhofer et al., 2014). Biomass production for use in electricity generation (hereafter biomass) is proposed as a renewable energy source that can contribute to the mitigation of climate change through direct CO2 sequestration and through the replacement of higher CO2 emitting fuels such as coal and oil (Bryan et al., 2008, Evans et al., 2010, Styles and Jones, 2007). The use of biomass (often in the form of agricultural residues, bagasse, forestry residues) is widespread globally, producing 280 TWh of electricity, equivalent to 1.5% of global electricity generation per annum (Eisentraut and Brown, 2012). But for biomass to play a significant role in future global energy supply, dedicated energy crops often grown on current agricultural land will be essential (Coleman and Stanturf, 2006, Evans et al., 2010).

Economically, biomass production has been found to be potentially competitive with conventional agricultural enterprises as the yields associated with production of woody perennials are often less sensitive to climatic variables and require fewer inputs (Bryan et al., 2010b, Heaton et al., 1999, Styles et al., 2008). In agricultural areas with variable climate and soil, the introduction of short rotation woody perennial production systems that use adapted woody species could provide a valuable diversification option. Moreover, it may offer the opportunity to buffer seasonal and annual variations in rainfall that cannot be reliably used by annual crops (Hobbs, 2009a). Internationally, where biomass supply chains are more developed, landholders have been slow in switching land use, particularly between agriculture and forested use despite potential profitability (Plantinga, 1996, Schatzki, 2003, Stavins and Jaffe, 1990). An explanation of this perceived investment inertia is that financial analysis of land use change has traditionally assumed the decision to switch land use can be modelled based on the Net Present Value (NPV) which compares current agricultural land uses with biomass alternatives (Yemshanov et al., 2015). However, several factors are commonly omitted from NPV analysis. Among them are sunk investment cost, investment irreversibility, significant uncertainty over future returns and flexibility in the timing of investment (Dixit and Pindyck, 1994, Trigeorgis, 1996). These omitted factors influence land holder decisions (Ihli et al., 2013) and lead to the erroneous NPV analysis conclusion that the land currently in agriculture would be more profitable in other forest based land uses (Frey et al., 2013, Parks, 1995, Stavins and Jaffe, 1990).

Real options analysis (ROA) has been proposed as a better model of investments decisions under conditions of uncertainty that are costly to reverse and where significant flexibility exists to delay investments (Dixit and Pindyck, 1994). ROA investment triggers, defined as the levels of revenue required to invest in a new land use, are often higher than NPV required returns if the investment involves inter-temporal opportunity costs (Musshoff, 2012). The effect of including ‘option values’ in investment decision analysis can be substantial (Regan et al., 2015, Schatzki, 2003). Unlike NPV analysis, the revenues required to trigger land use change must not only compensate the landholder for establishment cost and foregone returns from agriculture, but also for lost management flexibility and the revenue uncertainty from the new enterprise (Reeson et al., 2015).

Many of the key uncertainties influencing agricultural production such as rainfall, temperature and soil types vary spatially (Bryan et al., 2014). Heterogeneity of these factors has been widely included in NPV analysis in order to understand the spatial distribution of cost-effective land use change (Bateman, 2009, Bryan et al., 2010a, Crossman et al., 2011) which have found that landscape heterogeneity is likely to affect the location and timing of land use change. While qualitatively acknowledged, spatial variability has been largely overlooked in quantitative ROA of land use change. Limited exceptions demonstrating differing conversion threshold prices and conversion probabilities across space include Dumortier, 2013, Yemshanov et al., 2015 and Sanderson et al. (2016).

Another gap in the ROA of land use change is the effect of climate variability through time on yield. It has been shown to be the principal source of risk affecting long term economic viability of rain-fed agricultural systems in NPV assessments for semi-arid regions such as south east Australia (Kandulu et al., 2012). Despite NPV assessment showing that climate change is likely to provide landholders with additional production risks, surprisingly few studies have addressed the effect of climate change on agricultural land use change in a ROA framework. Hertzler, 2007, Hertzler et al., 2013 and Sanderson et al. (2016) are exceptions. They address these factors across an agricultural region with ROA employing spatial transects as an analogue for temporal changes due to climate change. There are limitations to this approach as temporal climate change effects are only roughly approximated by spatial transects. They exclude, for example, accounting for changing CO2 concentrations and their interactions with higher temperatures (Sanderson et al., 2016).

In this study we address both the gap in broad spatial coverage and the gap in accounting for climate change in ROA of land use changes. This study specifically modelled land use change from agriculture to biomass production in a spatially explicit framework across a broad region accounting for effects of climate change on yield variability. The analyses allow for the assessment of regional biomass industry viability with calculations and spatial mappings of areas where biomass land use is economically viable at several price points under alternative assumptions about climate change.

This article is organised as follows: The next section discusses the stochastic simulation-based real options model applied. This is followed by mapping the land use conversion to biomass with varying price and climate change assumptions. The final discussion focusses on how conclusions about regional biomass industry viability differ with ROA and traditional NPV analysis in the context of climate change futures.

Section snippets

Study area

Our study focused on the lower Murray region of southern Australia (Fig. 1). The dominant land use covering 50% of the region is rain-fed mixed farming, consisting of the dryland winter cropping of cereals (wheat, barley, oats), pulses (beans, lupins, peas), oilseeds (canola) and grazing of sheep (Bryan et al., 2011). The average farm size in the region is approximately 1000 ha (Kandulu et al., 2012). The region is typical of semi-arid rain dependant farming regions found globally. These

Regional primary productivity

Significant spatial heterogeneity of primary productivity was evident across the study area, determined primarily by rainfall and soil characteristics with greatest agricultural productivity in the high and medium rainfall areas. Modelling of both agricultural and biomass productivity suggest declines in both wheat and biomass production in response to warmer and drier climate futures in the low and medium rainfall zones (Fig. 3). But in the high rainfall zone, where winter crop production can

Discussion

The development of a large scale biomass industry in Australia has been suggested as a climate change resilient diversification option for agricultural landscapes. Currently, there are impediments to realising this potential. These include large reserves of accessible coal, the low cost of electricity generated in coal-fired power plants and uncertain greenhouse and renewable energy policy (Raison, 2006). Integrated tree processing is seen as a promising way forward for mallee based industries

Conclusions

The results presented here reaffirm that options values must be considered in land use change economics. However, the results demonstrate that the underlying variability in primary productivity has a substantial effect on the magnitude of uncertainty on the returns required to trigger land use change from wheat to biomass. Areas traditionally thought of as being quite similar on an average yield basis can display large differences in response to the inclusion of production and price risks due

Acknowledgements

This work was made possible by the Charles John Everard Scholarship awarded through the University of Adelaide and the support of CSIRO Sustainable Agriculture Flagship. The Authors wish to acknowledge Darran King and John Kandulu from CSIRO for proving spatial data sets and Matt Westlake and Waseem Kamleh from the University of Adelaide for assistance with high performance computing. The authors would like to recognise and thank the anonymous reviewers for their suggestions for improving this

References (85)

  • S.J. Jeffrey et al.

    Using spatial interpolation to construct a comprehensive archive of Australian climate data

    Environ. Model. Softw.

    (2001)
  • J.M. Kandulu et al.

    Mitigating economic risk from climate variability in rain-fed agriculture through enterprise mix diversification

    Ecol. Econ.

    (2012)
  • B.A. Keating et al.

    An overview of APSIM, a model designed for farming systems simulation

    Eur. J. Agron.

    (2003)
  • Q. Luo et al.

    Potential impact of climate change on wheat yield in South Australia

    Agric. For. Meteorol.

    (2005)
  • Q. Luo et al.

    Adaptation to climate change of wheat growing in South Australia: analysis of management and breeding strategies

    Agric. Ecosyst. Environ.

    (2009)
  • O. Musshoff

    Growing short rotation coppice on agricultural land in Germany: a Real Options Approach

    Biomass Bioenergy

    (2012)
  • P.J. Parks

    Explaining “irrational” land use: risk aversion and marginal agricultural land

    J. Environ. Econ. Manag.

    (1995)
  • R. Raison

    Opportunities and impediments to the expansion of forest bioenergy in Australia

    Biomass Bioenergy

    (2006)
  • A. Reeson et al.

    Management flexibility, price uncertainty and the adoption of carbon forestry

    Land Use Policy

    (2015)
  • C.M. Regan et al.

    Real options analysis for land use management: methods, application, and implications for policy

    J. Environ. Manag.

    (2015)
  • L.C. Rodriguez et al.

    Biomass assessment and small scale biomass fired electricity generation in the Green Triangle, Australia

    Biomass Bioenergy

    (2011)
  • T. Schatzki

    Options, uncertainty and sunk costs: an empirical analysis of land use change

    J. Environ. Econ. Manag.

    (2003)
  • R.E. Sims et al.

    Eucalyptus species for biomass energy in New Zealand—Part II: coppice performance

    Biomass Bioenergy

    (1999)
  • D. Styles et al.

    Energy crops in Ireland: quantifying the potential life-cycle greenhouse gas reductions of energy-crop electricity

    Biomass Bioenergy

    (2007)
  • D. Styles et al.

    Energy crops in Ireland: an economic comparison of willow and Miscanthus production with conventional farming systems

    Biomass Bioenergy

    (2008)
  • E. Wang et al.

    Modelling farming systems performance at catchment and regional scales to support natural resource management

    NJAS-Wageningen J. Life Sci.

    (2009)
  • M. Wolbert-Haverkamp et al.

    Are short rotation coppices an economically interesting form of land use? A real options analysis

    Land Use Policy

    (2014)
  • D. Yemshanov et al.

    A real options-net present value approach to assessing land use change: a case study of afforestation in Canada

    For. Policy Econ.

    (2015)
  • ABARES

    Australian Farm Survey Results 2010–11 to 2012–13

    (2013)
  • J. Antón

    Managing Risk in Agriculture: a Holistic Approach

    (2009)
  • J. Bartle

    Integrated production systems

  • J. Bartle et al.

    Scale of biomass production from new woody crops for salinity control in dryland agriculture in Australia

    Int. J. Glob. Energy Issues

    (2007)
  • I.J. Bateman

    Bringing the real world into economic analyses of land use value: incorporating spatial complexity

    Land Use Policy

    (2009)
  • M. Bennell et al.

    Evaluating Agroforestry Species and Industries for Lower Rainfall Regions of Southeastern Australia

    (2009)
  • B. Bryan et al.

    Lower Murray Landscape Futures - Data Analysis, Modelling and Visualisation for Dryland Areas

    (2007)
  • B. Bryan et al.

    Biofuels agriculture: landscape-scale trade-offs between fuel, economics, carbon, energy, food, and fiber

    GCB Bioenergy

    (2010)
  • B. Bryan et al.

    Influence of management and environment on Australian wheat: information for sustainable intensification and closing yield gaps

    Environ. Res. Lett.

    (2014)
  • N.D. Crossman et al.

    Carbon payments and low-cost conservation

    Conserv. Biol.

    (2011)
  • CSIRO

    Biofuel Database

    (2006)
  • R.H. Day

    Probability distributions of field crop yields

    J. Farm Econ.

    (1965)
  • A.K. Dixit et al.

    Investment Under Uncertainty

    (1994)
  • J. Dumortier

    The effects of uncertainty under a cap-and-trade policy on afforestation in the United States

    Environ. Res. Lett.

    (2013)
  • Cited by (25)

    • Carbon and water fluxes are more sensitive to drought than heat in terrestrial ecosystems in China

      2021, Journal of Hydrology
      Citation Excerpt :

      The energy partitioning of LE/Rn (net radiation) and H/Rn in terrestrial ecosystems presents considerable seasonal, interannual, and regional variability in previous studies (Eichelmann et al., 2016; Lei and Yang, 2010). A better understanding of changes in climate zones and PFTs in energy and water vapor exchange and their control mechanisms are essential for accurately predicting the energy and hydrological cycles in ecosystems and addressing the consequences of management decisions in terrestrial ecosystems (Meier et al., 2017; Regan et al., 2017). Climate anomalies and PFTs strongly influence the sensitivity of terrestrial carbon and water fluxes to substantial seasonal variations in climate factors (Angert et al., 2005; Eck et al., 2020; von Buttlar et al., 2018).

    • Adaptation to climate change: Extreme events versus gradual changes

      2021, Journal of Economic Dynamics and Control
      Citation Excerpt :

      Our paper also contributes to the growing literature on adaptation that uses real options models to capture uncertainties (Zhao, 2013; Zilberman et al., 2012). The previous literature has studied a variety of practical adaptation problems, ranging from flood risk management (Woodward et al., 2011; Gersonius et al., 2013) to investments in costal defense (Linquiti and Vonortas, 2012), water resource management (Jeuland and Whittington, 2014; Sturm et al., 2017), land use (Regan et al., 2017) and agriculture (Hertzler, 2007; Sanderson et al., 2015). These studies usually take the dynamic programming approach (Dixit and Pindyck, 1994; Copeland and Antikarov, 2003), but it is difficult to use this approach to handle situations when both gradual changes and extreme events are present.

    • A socio-economic exploration of landholder motivations to participate in afforestation programs in the Republic of Ireland: The role of irreversibility, inheritance and bequest value

      2020, Land Use Policy
      Citation Excerpt :

      Given the overall difficulty of converting afforested land back to agriculture under this scheme, the land use change is often perceived by farmers as irreversible and permanent (Howley et al., 2012). The irreversibility of investments in alternative agricultural land uses, including farm forestry has been acknowledged as an important factor potentially delaying or preventing investment and has been explored quantitatively in economics literature (Behan et al., 2006; Duku-Kaakyire and Nanang, 2004; Isik and Yang, 2004; Musshoff, 2012; Regan et al., 2015, 2017; Schatzki, 2003). Such research has sought to define a value that can compensate for the irreversibility of the investment based on market valuation of the decision.

    • The influence of crediting and permanence periods on Australian forest-based carbon offset supply

      2020, Land Use Policy
      Citation Excerpt :

      The selection of discount rate can vary with factors including the cost of capital, risk preferences/perceptions, time horizon of the investment and economic knowledge (Koppenberg and Spiegel, 2017). We test several discount rates (Table 1) found in the literature in order to capture the inertia commonly seen in adopting new land uses (Bryan et al., 2014; Dumortier, 2013; Regan et al., 2017; Thamo et al., 2017). Similarly to Thamo et al. (2017), the base case scenario adopted a 5 percent discount rate that approximates the real interest rate given the average long term cost of capital to small business in Australia of approximately 8.3 percent (RBA, 2019b) and average inflation rate of 2.8 percent (RBA, 2019a).

    View all citing articles on Scopus
    View full text